|
Bugzilla – Full Text Bug Listing |
| Summary: | Hal is broken and can't mount usb drives after installing scanner | ||
|---|---|---|---|
| Product: | [openSUSE] openSUSE 11.0 | Reporter: | Forgotten User XG9X5w8kVa <forgotten_XG9X5w8kVa> |
| Component: | Hotplug | Assignee: | E-mail List <bnc-team-screening> |
| Status: | VERIFIED NORESPONSE | QA Contact: | E-mail List <qa-bugs> |
| Severity: | Normal | ||
| Priority: | P5 - None | CC: | jsmeix |
| Version: | Final | ||
| Target Milestone: | --- | ||
| Hardware: | Other | ||
| OS: | Other | ||
| Whiteboard: | |||
| Found By: | --- | Services Priority: | |
| Business Priority: | Blocker: | --- | |
| Marketing QA Status: | --- | IT Deployment: | --- |
| Attachments: |
y2logs
/var/log/messages from oct 5 to oct 13 /etc/hal/fdi/policy/10osvendor/80-scanner.fdi |
||
|
Description
Forgotten User XG9X5w8kVa
2008-10-11 01:45:20 UTC
Please attach y2logs. If you are in doubt follow:http://en.opensuse.org/Bugs/YaST. Also if it's possible attach /var/log/messages after connecting a usb Thanks for report! Created attachment 246050 [details]
y2logs
Created attachment 246053 [details]
/var/log/messages from oct 5 to oct 13
I've cleaned the log of some unrelated events (sshd, susefirewall, dhcp, ntpd,...)
In this messages file you have plenty of normally working usb disk events at oct 05, before scanner installation.
The scanner was installed at 14:46 oct 8
I attached again usb disk devices at 02:19 oct 11, noticing then the hal problem; I tryed rebooting and with several usb devices.
At 3:16 oct 11 the usb disks mount again, after I reinstalled the hal package.
Reassigning back. Since YaST is causing the problem, the YaST developer should find out what they did (wrong) and how they influence HAL. It works for me and I never got a report like this. Please explain why "YaST is causing the problem"? What did you find out so that you know that YaST is the root cause? Because obviously everything worked before installing the scanner with YaST an it worked after reinstalling HAL. Looks to me clearly as if the YaST scanner installation changed something that caused the trouble. The only thing in the YaST scanner module which writes something regarding HAL is /usr/lib/YaST2/bin/test_and_set_scanner_access_permissions It writes /etc/hal/fdi/policy/10osvendor/80-scanner.fdi if the device is not already listed in /etc/hal/fdi/policy/10osvendor/70-scanner.fdi (the latter belongs to the sane-backends RPM). Miguel Angel Alvarez, please attach your /etc/hal/fdi/policy/10osvendor/80-scanner.fdi file. Created attachment 248408 [details]
/etc/hal/fdi/policy/10osvendor/80-scanner.fdi
/etc/hal/fdi/policy/10osvendor/80-scanner.fdi, but current file, after hal reinstall.
Yast scanner module is having quite a weird behaviour. I uninstalled the scanner and removed iscan-free to try to reproduce the issue, deleted 80-scanner.fdi too, but the issue didn't reproduce. Instead, after the first reinstall, I got a dialog message after installing iscan-free: Warning. Coudn't set scanner access permisions. Error message is: Cannot execute /usr/bin/hp-makeuri. The following models are currently not known to HAL: USB-ID(hex)=04b8:010c. To access the scanner as normal user, udev, HAL, and hal-resmgr are needed to grant appropiate access permissions automatically. Therefore the scanner model must be known to HAL. If the scanner is not known to HAL, a replug of a USB scanner should help. Otherwise a reboot should be done..... After a new uninstall and reinstall, that message didn't appeared. The yast scanner module hangs with y2base eating 100% cpu if I press the ADD button. Killing y2base brings a dialog with the following message: YaST got signal 15 at YCP file Wizard.ycp:740 /sbin/yast2: line 421: 6002 Terminado $ybindir/y2base $module "$@" "$SELECTED_GUI" $Y2_GEOMETRY $Y2UI_ARGS Attachment #248408 [details] looks perfectly o.k. for me. It has no real entry for HAL at all (only a comment) which is as I expected it because your scanner model Epson Perfection 640U with USB IDs 0x04b8:0x010c is already listed in /etc/hal/fdi/policy/10osvendor/70-scanner.fdi so that the YaST scanner module does not add it to /etc/hal/fdi/policy/10osvendor/80-scanner.fdi. Both files do not belong to the hal package so that reinstalling of the hal package would not have any effect regarding those files. The first paragraph in comment #9 shows that there are indeed whatever problems in the HAL layer or perhaps in lower level layers e.g. udev or in the kernel USB subsystem. You can ignore the "Cannot execute /usr/bin/hp-makeuri" part because it is needed only for HP all-in-one devices (/usr/bin/hp-makeuri belongs to the hplip package). In particular when a model is already listed in /etc/hal/fdi/policy/10osvendor/70-scanner.fdi it works normally out of the box without the need for any reboot. The YaST scanner module only reports that there are possible problems (i.e. no bug in YaST). The second paragraph in comment #9 is a bug in the YaST base GUI system which needs very very much time when it should display a so called SelectionBox or a Table with many entries. In your case after clicking "Add" you would get a list of all known scanner models (several hundred up to about two thousand). This is also no bug in the YaST scanner module (but in the YaST base GUI system). In this bug report please focus only one the HAL problem to avoid that a report gets mixed up with different issues which have nothing to do with each other. Miguel, can you try this out: 1. get root user or use sudo 2. open /etc/init.d/haldaemon in an editor of your choice and change: HALDAEMON_PARA="--daemon=yes"; to HALDAEMON_PARA="--daemon=yes --verbose=yes --use-syslog"; 3. call: logger =============================== 4. then restart HAL via: rchal restart 5. check if hal is ready by calling lshal and checking if device information get printed 6. reproduce the problem 7. attach the part of /var/log/messages since the =============================== in the syslog to the bug. Regarding the issue that the YaST base GUI system needs very much time to display long lists see bug #442173. If you can provide the requested information, please reopen this bug. |